You may have saved for decades and invested your money carefully. But when the money must start flowing in the opposite direction – from your portfolio to you – you are suddenly faced with a whole new set of challenging choices.   

This article is built around six tables of numbers that should be of interest to anyone who is retired or who is planning for retirement. These tables are extremely useful to us when we work with clients.

Before we dig in, let’s back up for a moment. I mentioned four major financial decisions that will shape your future once you retire. In addition to the two we are examining here, the other two are:

How will you invest your money?
How much risk will you take with your investments?

These last two questions are related. You’ll find our recommendations for how to invest, along with the reasons for those recommendations, in an article called “The ultimate buy and hold strategy.”  That article tells you what kind of stocks and what kind of bonds are most likely to give you the best long-term returns, based on all the history we can get our hands on.

However, that article leaves out one very important thing. It doesn’t tell you how much of your portfolio should be in stocks and how much in bonds. That’s related to risk, which is the topic of another important article called “Fine tuning  your asset allocation.”

Taking money from your portfolio

Back to the topic at hand. Only you can answer the questions of how much you need or want from your portfolio and whether you can handle a variable income instead of a fixed one. Every situation is different, and you will probably get the best answers with the help of a professional financial advisor.

Without knowing all your circumstances, I cannot give you the answers that are best for you. However, in this article I will give you a head start in thinking about the questions, something that should be extremely helpful preparation for a meeting with an advisor.

This article has been a favorite of readers on our web site for quite a few years. (It used to be called “Retirement: When your portfolio starts paying you.”) You are reading a 2010 update of a major revision that we published in the spring of 2009.

Because of investors’ concern about future inflation, the distributions shown in three of the six tables have been adjusted for actual inflation that occurred during the years in this study. Prior to 2009, we simply assumed a steady rate of inflation – an approach that’s simple on paper but does not reflect real life.

This article describes three possible withdrawal rates: 4 percent, 5 percent and 6 percent. We start with 5 percent, a good ballpark figure for many people. However, 5 percent may be too conservative for some investors and too aggressive for others. 

The tables that go with this article show returns of globally diversified portfolios that are built with Dimensional Fund Advisors mutual funds. These portfolios are similar but not identical to those we manage for clients. Returns are net of assumed management fees of 1 percent annually plus applicable transaction costs. Percentages in the top row indicate various combinations of equities and fixed income.

Fixed income vs. variable

A major decision you make when you retire is whether you will choose a fixed withdrawal plan or a flexible one. I'll start by describing the fixed withdrawal plan.

For the purposes of this article, I’m going to assume three things:

•    You retire with a portfolio worth $1 million;
•    You need to supplement your other income (Social Security, pension, rental income and so forth) by taking $50,000 from your portfolio the first year;
•    You will need to adjust that annual withdrawal every year to keep you protected from inflation.

We will refer to this plan as a fixed distribution schedule. Although the withdrawal each year will change, you will maintain a fixed amount of spending power each year as determined by changes in the Consumer Price Index.

Studying the numbers

Table 1 shows the results of doing this, starting with 1970, the earliest year for which we have full data.

If you’re not used to looking at a table like this, here’s a quick guide. On the far right side is a column showing the actual distribution every year, which is derived from actual inflation in the previous year. For example, inflation in 1970 was 5.66 percent, and that raised the 1971 distribution to $52,830.

The other columns show the year-end values of the portfolio for various asset allocations. The portfolios begin on the left with the most conservative and become more aggressive as you move to the right.

You’ll notice immediately that there’s some white space at the bottom of six of those columns. The reason is simple: those six portfolios ran out of money under these assumptions.

For example, if you had invested exclusively in the Standard & Poor’s 500 Index, by the end of 1991 your portfolio would have been worth only $182,041 (down from $1 million to start) and just barely able to pay you the $177,604 you needed in 1992 to match the spending power of the $50,000 you took out in 1970. After that, there was almost nothing left.

You’ll see that the more conservative portfolios on the left also ran out of money at various points. Six of the globally diversified portfolios held up all the way through these 40 years (longer than most people’s retirements), and five of those six had plenty of assets remaining at the end of 2009. 

The obvious conclusion is that, at least for the pattern of returns and inflation in these years, a prudent investor needed to have at least 50 percent of his or her portfolio in equities. 

A balancing act

Risk and return are a balancing act, and Table 1 shows it. Investors who were unwilling to risk having much of their portfolios in equities eventually wound up taking the greater risk of running out of money. On the other hand, investors who sought higher long-term returns by owning more equities had to somehow keep their faith during the first few years of retirement. That was not a piece of cake.

Note that the 100 percent global equity portfolio dropped in value to $740,423 at the end of 1974. That had to be frightening for retirees who had started with $1 million. In hindsight, we can see that it all worked out quite well from that point forward. But there was no way to know that in January 1975 as the plan called for removing nearly $70,000 from that portfolio.

Another thing that jumps out at me from Table 1 is the cumulative effect of inflation. It’s very easy to dismiss inflation as a minor financial force, but as the distributions column in this table shows, inflation can be very powerful. What started out as a “mere” $50,000 in 1970 became $101,810 only 10 years later.

Before leaving this table, let me point out one other pretty interesting number that you’ll find at the bottom, labeled “Total Distribution” – nearly $6.5 million. That figure means the six portfolios that survived all these years paid out about $6.48 for every $1 that they started with.

In addition, if you adjust the portfolio values at the end of 2009 for cumulative inflation, the globally diversified portfolios with allocations of 60 percent or more began 2010 with more spending power than they had at the start of 1970.  Not bad at all, considering all the income they had generated in the meantime.

(Future returns will be different from those shown here, and inflation will be different. So don’t take these numbers as results you can expect.)

Flexible distributions

Now that you know how to read this table, I’d like to move on to another dimension: a flexible distribution plan. I have long believed that one of the ultimate financial luxuries in life is to have saved enough money to take flexible distributions in retirement.

A flexible distribution is one that changes, but not according to inflation the way we have shown in Table 1. In this case, the yearly distribution goes up and down according to the value of the portfolio. In other words, it automatically does what most smart retirees would naturally want to do if they could: Take out more money after good investment returns and scale back on withdrawals when the portfolio is suffering.

Recall for a moment the first five years in Table 1. The retiree who followed that plan got more money every year, regardless of what the stock market was doing. His withdrawals were insulated from the big bad bear market of 1973 and 1974.

Table 2 shows a flexible distribution plan that also started out with a $50,000 withdrawal in 1970. But this time, subsequent distributions began with a drop, reflecting the market. Eventually the flexible distributions regained that $50,000 level, but in the meantime our theoretical retiree had to live on less real money than he had started with.

The layout of Table 2 is slightly different from that of Table 1, because each year’s distribution was different for each portfolio allocation.

You can see that in 1975, the distribution in each column was at its low point, reflecting the stock market’s dismal performance in the previous two years. From there, they went up, though gradually.

By 1987, flexible-plan distributions (Table 2) were higher than fixed-plan ones (Table 1) in the 50 percent and 60 percent global equity portfolios – and they stayed ahead from that year forward.

However, for the second through 17th years of retirement, investors in the flexible plan had to make do with less purchasing power than they had in 1970.

You call that an ultimate luxury?

You may be wondering why anybody would be willing to embark on a flexible plan like this. It’s a very good question.

Recall what I said earlier: I have long believed that one of the ultimate financial luxuries in life is to have saved enough money to take flexible distributions in retirement.

For example, imagine that you retired needing $50,000 a year from your portfolio, but instead of $1 million, your portfolio was worth $1.5 million at the start of 1970. In that case, every distribution would be 1.5 times as great as shown in Table 2.

Your first-year distribution would give you 50 percent more money than you really needed, allowing you to spend money on some of the extras you’re likely to desire in your first year of retirement.

Using the 60 percent equity globally diversified portfolio for an example, the low point of your distributions would be (again in 1975), $66,438 – or 1.5 times the $44,292 shown in Table 2. Table 1 tells us that in 1975 it required $68,891 to match the spending power of $50,000 in 1970. If all you could spend that year was $66,438, you would have had to tighten your belt a little bit. But not a great deal.

And two years later, in 1977, you’d have $91,299 to spend in the flexible plan. That’s comfortably above the $77,254 that it would have taken to keep up with your $50,000 cost of living plus inflation. And from that point forward, your flexible distributions would keep going up nicely.

At least as important as that, your portfolio would have easily survived for 40 years, longer than most retirees are likely to live. Remember, this seemingly abundant retirement was possible because you had saved 1.5 times as much as you really needed.

Fixed vs. variable

Here’s my take on fixed vs. variable. If you have saved more than enough money to meet your needs with an inflation-adjusted withdrawal rate, you may be able to afford a flexible distribution plan in which what you take out of the portfolio is determined by your investment results instead of by inflation.

Based on the years in this study, this could give you peace of  mind knowing you are less likely to run out of money. It may leave more for your heirs. And it is likely to give you more spending power at some point in your retirement years. However, exactly when that happens (and whether it happens at all) is entirely dependent on patterns of market return that are highly unpredictable.

My advice therefore is to save more than you think you will need. I doubt very much that I will ever get a call from you complaining about the awful trouble you are in because you took my advice and saved too much money. If I do get that call, we will have an interesting conversation.

Take out more? Take out less?

There’s nothing cast in stone about taking 5 percent from a retirement portfolio every year. As we saw in Tables 1 and 2, that withdrawal rate would have worked well for many combinations of assets in the period starting in 1970. And each combination in Table 2 ended 2009 with more purchasing power than it started with 40 years earlier.

Many people want to take out more. So I’m going to show you the hypothetical result of that in Table 3, based on a $1 million starting portfolio value and an initial withdrawal of $60,000.

Other people, for various reasons, may want to be more conservative and take out less. You will find that in Table 4, based on an initial withdrawal of $40,000, or 4 percent of the initial $1 million. 

Before you look at more of these tables, can you guess what they will look like? If you withdrew money at a rate of 6 percent instead of 5 percent, would you expect the portfolio values after a few years to be more than in Table 1? Or less?

Right! The portfolio values would logically be lower after some years of taking out more. And that’s just what you see in Table 3 . In fact, you can see that only three of those portfolios survived all the way to the end of 2009. And one of those three, the 80 percent global equity mix, obviously was in its last days; it had less than $1 million left from which to pay out more than $300,000 a year.

That leaves only the 90 percent and 100 percent columns in the long-term running. Both of them represent portfolios involving more risk than I think is appropriate for most retirees. Remember, this is a plan for people who really need that $60,000 a year and who don’t have much tolerance for failure.

For people with relatively short life expectancies and no burning desire to leave money to their heirs, this might work. But for most people, I don’t think it’s the right plan.

Now turn to Table 4 , in which (as you no doubt guessed without looking at it) almost all the portfolios held up very well for many years. The difference is the low withdrawal rate, akin to “sipping” from the portfolio instead of “drinking” or “gulping” from it. Over 40 years, all the portfolios with 30 percent or more in equity, even the undiversified Standard & Poor's 500 Index, survived just fine.

For investors who can meet their needs while taking out only 4 percent of their assets and who want to leave significant assets to their heirs, this could be a very desirable plan.

Table 5 and Table 6 show the results from flexible distribution plans at 6 percent and 4 percent. You now know how to read these tables, and you can draw your own conclusions. Again, I believe that flexible distributions are most appropriate for people who have over-saved.

If you assume starting portfolios of $1.5 million instead of the ones shown here, you can multiply each distribution by 1.5 and see that in most cases those retirees would have had a good ride.

Lessons from these numbers

If this discussion leaves you with only one lesson, I hope it is the value of having more money, instead of less, when you retire.

Of course we don’t always have a choice about when we retire. In those cases our resources are whatever they are, and our challenge becomes making the most of them. These tables will help you think about how to do that, based on your own circumstances.

If  you are ready (or think you’re nearly ready) to retire, the tables in this article may help you form a general idea of what your retirement could look like financially.

One simple way for many people to improve their financial outlook in retirement is to work a few extra years.

This has at least four important benefits. First, additional years on the job will let you add more to your savings. Second, your portfolio has more time to potentially grow before it has to start paying you. Third, after your retirement your portfolio will have fewer years it must make payouts to you – meaning those payouts can be larger. Fourth, if you delay taking Social Security, your payments will be permanently higher.

Finally, if you’re a young person with many years before you plan to retire, I hope you’ll consider ramping up your savings plan, now that you know more about how retirement income really works.

Paul Merriman is founder of Merriman.